At AGBT 2016, the Winners Are Long Reads and Whole Solutions

AGBT_logo_RGBThis year’s pilgrimage of geneticist, genomics scientists, and along with that commercial companies was to the JW Marriott in Orlando and not as usual to Marco Island – however, we were assured that the next two AGBT events are already confirmed to be held on Marco Island again. The Orlando Marriott setting seemed fine; nothing beats the Marco island environment with its beach setting which is well worth the extra hour bus ride from the airport to take you there. Having said this, the setting appeared to be more favorable for the vendors in particular, given the more open arrangement of large suites inviting easy access and interactions with conference attendees.

In general, this year’s meeting followed the true and tested script of past installments. Interestingly this year, as the technological advancements seemed to have slowed and somewhat resulting in fewer breakthrough announcements, a larger emphasis was on whole solutions and the long read technology or partnership with 10x Genomics (see details later in this post). This was readily apparent with the focus on whole complete solutions such as the Qiagen’s GeneReader NGS System “Sample to Insight”, Roche’s investment in the PacBio Sequel technology for clinical structural variant detection, or Illumina’s vision on the Firefly project (discussed in more details below). Software analysis announcements around secondary and tertiary sequence data analysis solutions were also on the sparser side.

The poster session was conveniently arranged in one location with the software demo setting which allowed for better exposure of the different tools. Some of the commercial tools demoed included secondary and tertiary analysis solutions, LIMS systems and workflow management solutions, public databases from commercial providers like Bina Technologies (a Roche company), DNAnexus, Lab7 Systems, Genologics (an Illumina company) Omicia, Qiagen, Seven Bridges Genomics, Station X, and others.

Clearly, the trend for companies seems to announce major updates and product news at the JP Morgan conference in the beginning of the year turning the AGBT conference into more of a follow-up conference (see also my blog JP Morgan Healthcare conference from Afar with a list of key announcements). Here, I am summarizing some of the news that accompanied this year’s AGBT conference, plus a couple of talk highlights. To read up on additional AGBT content check out the AllSeq AGBT 2016 blog of blogs or the blog rolls of the official AGBT bloggers Dale Yuzuki, Meredith Salisbury, and James Hadfield.

Talk highlights

I decided to only highlight a few talks within this post as there is overlap with some talks presented at PMWC 2016 which I summarized in my most recent blog PMWC 2016 Silicon Valley: Scalable Clinical Data Interpretation The Biggest Bottleneck.

  • Jay Flatley provided an overview of the announcement made at JP Morgan. New this week is the onboarding of the new CEO of Grail, Jeff Huber who joins from Google where he was Senior Vice President. Grail is developing an early cancer detection blood test for patients showing no symptoms at all. Some additional details on the Firefly project:
    • Firefly: Though only to be released in late 2017 Jay Flatley provided an update on this semiconductor sequencing platform, which is based on the CMOS technology (acquired in 2008 with Avantome). Considered to be simple from “samples to answers” the goal is to allow blood as sample input and a final report as analysis outcome – clearly following the latest trend of making solutions as simple and complete as possible. Firefly’s chemistry will use a new encoding technology. In its four-channel chemistry, which is used by Illumina’s HiSeq family of instruments, each nucleotide is labeled with a separate fluorescent dye that is detected in four different optical channels. Flatley stated that their goal is to have application developers build solutions for this open platform which is why Illumina is so public about this release this early in the process. Some stats includes:
      • 99% raw read accuracy similar to the HiSeq system
      • 5 – 13 hours run time with 1gigabase per run output (solution for targeted discovery and PGS)
      • all data will be directly loaded into BaseSpace
      • target is the hospital setting
      • cost per sample will be ~$100 with the entire system below $30K
  • Anne Wojcicki provided an exciting update on 23andMe. To date, 23andMe has about 1.2 million customers with over 80% of them consented to join the discovery aspect of the company – she calls this an engaged research cohort. Wojcicki updated on the IRB approved process which allows all customers to opt out at any time. Interestingly 75% take at least one survey, which allows 23andMe to collect about 2 million data points (individual survey responses) every single week, amounting to a total of over 320 million phenotypic data points collected to date. As a result, these customers contribute to over 230 different studies (e.g. Parkinson study). Wojcicki stated that this huge customer engagement factor is a direct result of returning the findings back to the customers. As such over a three-year period about 42% have logged into the system at least once within a 90 day period (or 38% after six years). Wojcicki noted that “providing findings in black and white back to an individual helps engaging them” as “genetic information can be a catalyst that something can happen based on a genetic profile, but also that something can actually be done about it (e.g. behavioral change).
  • A very energetic and fascinating talk was given by Chris Mason (Weill Cornell ScottKellyMedical College) about the NASA Twins study. Mason introduced the study “going from a moon shot to a mars shot”! The goal of the study is to investigate insights into the broader effects of lack of gravity by studying two individuals (Scott Kelly in space and his twin brother Mark Kelly on earth) for a period of one year. The study includes profiling of the two probands during as well as 6 month before and after the one year endeavor, which includes studying the microbiome, the DNA methylation pattern, clinical chemistry values, and expression and variant analysis. They are taking the route of collecting the samples and “shipping” them back to earth for analysis both in the Mason and Snyder lab (the entire study though includes a total of ten separate investigations). At this point the researchers have already identified some differences of interest, but none of those are revealed until the study will have concluded. Mason also briefly touched base on the “terrestrial protocol” he is working on which includes sequencing in space with the MinION sequencer.

Announcements accompanying the conference

10x Genomics and their long read application

  • Illumina announced long read application partnership with 10x Genomicsto enable long read applications and co-marketing to promote 10x Genomics Linked-Read sequencing products and Illumina sequencing systems. This will enable Illumina customers to effectively complete projects that require phasing, structural variant analysis, de novo genome assembly and the remapping of difficult regions of the genome by partnering with select companies. The companies will co-market the 10x GemCode technology and Illumina sequencing solutions to their respective customers.
  • Agilent Technologies and 10x Genomics Announce Collaboration to Develop a Premium Exome allowing in-depth discovery of important novel content within and around the exome by enabling phasing, access to previously unmappable loci, structural variation, and copy number detection through the use of complementary products of both companies. The two companies will each create solutions that jointly provide a streamlined workflow for 10x Genomics’ Chromium platform using Agilent’s market-leading SureSelect target enrichment technology. The collaboration entails coupling the 10x Genomics Linked-Read technology with an optimized Agilent SureSelect target enrichment solution to deliver coverage and variant calling, particular in hard-to-map regions of the genome.
  • 10x Genomics introduced the new Chromium System at their workshop. Results and multiple applications were demonstrated by collaborators at the Broad (Stacey Gabriel) and the University of Washington (Scott Furlan) using the GemCode™ technology.

Cloud applications

  • Spiral Genetics announced a cloud partnership with Microsoft to make their respective bioinformatics platform available on the company’s Azure cloud. Specifically, Spiral Genetics plans to deploy its BioGraph, a method of compressing and querying large quantities of sequencing data (developed in collaboration with Baylor College), in the cloud.

Making use of consumer genetics data

  • 23andMe and Celmatix Collaborate to Improve Infertility Outcomes This collaboration will also enable the development of early screening tests which will help OBGYNs identify women who are at risk for premature decline of their ovarian function, which has both infertility and broader health implications. The collaboration will combine Celmatix’s discovery and annotation of more than 5,200 genetic biomarkers related to fertility potential in humans and 23andMe’s genetic database and research community.

Next-generation sequencing

  • QIAGEN Launches Targeted RNA Panels for Next-generation Sequencing Sample to Insight workflows. More than 170 new QIAseq Targeted RNA Panels, which cover an extensive range of disease- and signaling pathway-focused genes (~100-500 genes per panel) for gene expression profiling, were released which enables researchers to select from over 20,000 human genes and lncRNA to survey expression fold changes and discover interactions between genes, cellular phenotypes and disease processes. The new QIAseq Targeted RNA Panels use molecular barcode technology and built-in control assays.
  • BD launches the NGS library prep BD CLiC™ System which enables genomic research by providing cost-effective NGS library preparation via a high-throughput, fully integrated, next generation sequencing (NGS) library prep instrument, engineered for targeted and whole genome library preparation. The BD CLiC consolidates the entire NGS library preparation workflow process into a single instrument.

Big data – cohorts

  • Genomics England announced a partnership with Illumina to develop a platform and knowledge base to improve and automate genome interpretation. The tools will operate within the Genomics England secure database to enable researchers and clinicians to readily access information and reports. This non-exclusive partnership with Illumina will run in parallel with the other clinical interpretation and bioinformatics providers involved in the 100,000 Genomes Project. Illumina and Genomics England will collaborate to develop a set of informatics tools, which will support the delivery of genomic clinical and research services at a population scale to the NHS Genomic Medicine Centers and the Genomics England Clinical Interpretation Partners. All of the tools will include open application programming interfaces so that other bioinformatics solution partners can continue to provide services within the Project.
  • GenomeAsia 100K Initiative Announced to Sequence 100,000 Genomes in South, North and East Asia to promote genetic understanding in support of research and discovery. The initial intent is to include populations from 12 South Asian and at least 7 North and East Asian countries. In the first phase, the project will focus on creating phased reference genomes for all major Asian ethnic groups representing a major step forward in understanding the population history and population substructure of the region. The sequencing of 100,000 individuals will be combined with microbiome, clinical and phenotype information to allow deeper analysis of diseased and healthy individuals in the context of inferred local ancestries.


  • UPMC Gene Test Brings Personalized Medicine to Cardiac Care. Clinicians at the University of Pittsburgh Medical Center are bringing personalized medicine into the routine care process for cardiac patients in need of stents for clogged arteries by using a simple blood test to identify gene variants that may impact their treatment.